Challenging Japanese Stock Swing Trading with ChatGPT #43
I continue to engage in short-term swing trading, where I purchase Japanese stocks during my weekday lunch break and, in principle, sell them during the next business day’s lunch break. I use ChatGPT for stock selection and trading decisions, and I am verifying with my own money whether it is actually possible to make a profit in Japanese stock trading using AI.
This article is also written by ChatGPT, with the exception of the final “Thoughts on this session” section. The main text is organized based on actual trading records and the analysis performed at that time.
A profit of 11,710 yen from FANUC. Despite a total profit of +10,710 yen across 3 stocks, the cumulative profit/loss remains in the red.
In the trades on October 28, I sold the three stocks I had purchased the previous day, locking in a total profit of 10,710 yen. While the result was 2 wins and 1 loss, the majority of the profit came from FANUC.
Results of purchases on October 27 and sales on the 28th
FANUC (6954)
I purchased 100 shares at 5,782.9 yen on October 27 and sold them at 5,900 yen on the 28th. The realized profit was +11,710 yen. This is an increase of approximately 2.03% from the purchase price.
Sumitomo Pharma (4506)
I purchased 100 shares at 1,937 yen and sold them the next day at 1,938 yen. The realized profit was +100 yen. Although the stock price trend was upward, the rate of increase was limited to approximately 0.05%.
Fujifilm Holdings (4901)
I purchased 100 shares at 3,457 yen and sold them the next day at 3,446 yen. The realized loss was -1,100 yen. The rate of decline was approximately 0.32%.
The total for the three stocks is +10,710 yen. All figures are profit/loss before the deduction of taxes and transaction fees.
Was ChatGPT’s judgment from the previous day correct?
For FANUC, the main evaluation factors were the revision of earnings forecasts announced on October 23 and a share buyback program with a limit of 50 billion yen. In the previous day’s analysis, I also confirmed the stock price movement above the VWAP and a recovery from a short-term dip, making it the highest-rated of the three stocks.
Even at the time of sale the next day, it was about 2% above the purchase price, so the hypothesis that an increase accompanied by news would continue until the next business day’s lunch break held true for this trade.
On the other hand, for Sumitomo Pharma, I evaluated the fact that the drug under development, ulixertinib, had received Breakthrough Therapy designation from the US FDA. However, the designation does not mean approval, and there were concerns about overheating or profit-taking after the announcement of the news. The result was a profit of only 100 yen, and the expected sustainability of the rise could not be confirmed.
For Fujifilm Holdings, I evaluated the resilience near the VWAP and the increase in trading volume, but there was no strong new information. The sale price the next day was lower than the purchase price, so the judgment expecting a short-term rebound was incorrect.
Also, the previous day, ChatGPT had set the purchase limit for FANUC at 5,780 yen and for Fujifilm Holdings at 3,449 yen. The actual execution prices exceeded the limits in both cases. While FANUC rose this time, the effectiveness of trades that do not adhere to price conditions needs to be verified separately.
On October 28, I newly purchased TDK and Shin-Etsu Chemical Co., Ltd.
On this day, I purchased two new stocks.
**TDK (6762)**: Purchased 100 shares at 2,768.8 yen, for a total investment of 276,880 yen.
During the morning session, the price recovered from a low of 2,736.5 yen and returned to near the VWAP. While there were positive factors such as a target price increase, the stock was already up about 5% at the opening, and ChatGPT was cautious about buying at a high. It was not an unconditional buy recommendation.
**Shin-Etsu Chemical (4063)**: Purchased 100 shares at 6,066 yen, for a total investment of 606,600 yen.
In the earnings report released the previous day, first-half recurring profit increased by approximately 13% year-on-year. Trading volume in the morning session was also significantly higher than usual, confirming the market’s reaction to the earnings data.
However, the momentum of the rise after the opening slowed, and ChatGPT had set a condition to avoid follow-up buying at prices exceeding 6,060 yen. The actual execution was 6 yen above this.
Cumulative 115 trades. Next, I will also verify compliance with purchase conditions.
The cumulative performance including this settlement is 62 wins, 53 losses, a win rate of 53.9%, and a realized loss of -143,000 yen.
A feature of this time is that while the decision to evaluate corporate materials for Fanuc led to profit, I was unable to secure sufficient price range for Sumitomo Pharma and Fujifilm Holdings.
Furthermore, there have been continued instances of purchasing above the price limit set by the AI. Moving forward, it is necessary to verify not only the accuracy of stock selection but also the difference in profit and loss between cases where purchase price conditions were met and where they were not.
On the afternoon of October 29, I plan to sell TDK and Shin-Etsu Chemical. Whether the earnings materials and the morning recovery will continue until the next day, and whether purchasing in a high price range will lead to profit, will be the verification points for next time.
Thoughts on this session
・Things have been going well lately, but when I look at the grass on the other side, like Kioxia, it looks greener. There was a rumor that Kioxia could never break the 12,600 wall, so if I had touched it yesterday, it would have been a margin short sale.・Shin-Etsu Chemical is the same pattern as Fanuc, or rather, the second day after good earnings pattern, so it would be nice if it goes up tomorrow, but looking at today alone, it’s quite a high-price buy.・Fujifilm was featured so much in Nikkei articles, why didn’t it go up?・I don’t know how long the winning streak will continue, but judging by the PTS, it might end today. I want to end October’s realized losses in the five-digit range!
※This article is a personal trading and verification record and does not recommend buying or selling specific stocks.
#Investment #StockInvestment #JapaneseStocks #Stocks #AssetManagement #AssetFormation #SwingTrading #InvestmentRecord #IndividualInvestor #StockInvestmentBeginner #ChatGPT #AIInvestment #TechnicalAnalysis #CorporateAnalysis #InvestmentStrategy